Managing paddocks more effectively

Figure 1. Protein map derived from a barley field, near Dalby, in 1999

By Rob Kelly, Wayne Strong, Troy Jensen, David Butler, Natasha Wells (QDPI&F) and Armando Apan (USQ)

Northern graingrowers, like those in other states, have long observed spatial variability within crops, prompting questions such as “Can I identify under-performing areas within my paddock and if so, can I manage parts of the paddock more cost-effectively?”

Alternatively, some growers may have asked whether it is possible to identify areas of different grain quality before harvest, to help better segregate grain during harvest. The advent of yield-monitoring devices have further prompted growers to manage some paddocks according to spatial variation where the variation is consistent from crop to crop.

Since 1999, three projects jointly funded by the GRDC and Queensland DPI&F have delved into the capture of spatial paddock information and its application to improve cost-effective land and/or crop management in the northern region.

In the first project, our prime objective was to increase the value of yield mapping for future crop management. We discovered that when a map of grain protein variation is used with the yield map, the management of nitrogen (N) inputs for future cereal crops is more cost-effective.

On-the-go protein measurement at harvest using a protein monitor is not yet fully commercially available. Nevertheless, coincidental monitoring of yield and protein would enable retrospective assessment of the crop"s N nutritional status. Collective information gathered from the yield and protein maps enables “N supply” (kg/ha) available during the growth of that crop to be estimated.

Factors learned from results of hundreds of multi-rate N fertiliser experiments in the past 30 years with wheat, barley or sorghum crops are used to estimate the quantities of available N needed to produce a grain N (kg/ha) offtake, for grain of a particular protein percentage.

The major drawback to the use of this spatial information is that because it is available after crop harvest it can only improve management of future cereal crops.

Remote sensing promises a similar opportunity to identify areas of nutritional or other crop stress, with the advantage of enabling management decisions before harvest.

Satellite and aerial spectral sensors that capture in-season spectral information (reflected off crop canopies) could allow interventions to crop management, provided the spectral information used is a surrogate measure of crop yield, grain protein, or of a crop stress responsible for creating variations within these attributes.

In a second project completed in 2003, imagery derived from satellites and other aerial platforms, including aircraft and balloons (see pictures), showed promise in identifying areas in cereal crops of high and/or low grain protein. This capability would be invaluable for better grain segregation at harvest. An equally valuable use for this information would be improved forecasting of grain production by protein classification to secure an early marketing advantage (Fig. 1).

Figure 1. Protein map (far left) derived from a barley field, near Dalby, in 1999. The Landsat-5 image(left), obtained in mid-September, displayed a similar pattern to grain protein harvested in early December (r2=0.71).

Part of our contribution to the GRDC"s Strategic Initiative on Precision Agriculture (SIP09) will be to pursue applications of remotely sensed spectral data that enable improved land and crop management decisions.

Spatial (or pixel size) and spectral resolutions (the number of spectral bands) of remotely sensed imagery are being improved as technology adapts to agricultural requirements. Images captured from satellite, aircraft or balloons can be taken at repeated intervals, allowing for timely data acquisition and near-real-time remedial action.

Spectral data from satellite images have been used in Australia and elsewhere to forecast production, locate and identify foliar diseases, aid in irrigation scheduling and to enhance fertiliser management.

If forecasts were available by protein classification, such advice could provide valuable marketing advantage for the producer and/or grain marketer.

As well as providing opportunity to segregate grain by protein class at harvest, remotely sensed imagery during crop growth will provide early detection of areas of under-performance within the paddock. Identifying the underlying cause for under-performance by systematic "ground-truthing" could lead to intervention, avoidance and/or corrective strategies. With Dr Apan (University of Southern Queensland), Dr Stuart Phinn (University of Queensland), and David Lester (Incitec/Pivot), we have linked canopy reflectance with yield-deficient supplies of sulfur, phosphorus and N in wheat.

Identification, avoidance and/or management of any soil dysfunction, including soil-borne pests (such as nematodes) or diseases that may affect sustainability of northern crops should improve production.

Tools such as this could greatly facilitate the collection of information useful for crop auditing or monitoring critical growth stages. An example of such an application for remotely sensed imagery in this region is to identify areas where subsoil constraints (sodicity, salinity, unfavourable pH or compaction) may restrict water use by dryland field crops.

For more information:
Dr Rob Kelly, 07 46881524, rob.kelly@dpi.qld.gov.au

GRDC Research code: DAQ 00067

Region North